The proposed study will develop and evaluate "Tools for Inpatient Monitoring of adverse Events (TIME) for safe and appropriate testing." Our research hypotheses are: (H1) A therapy-lab result DB will enable determination of the time course, severity, and co-determinants of adverse effects (as measured by clinical laboratory results) of various therapies (i.e., the database can provide an evidence-based mechanism for documenting therapy-lab effect pairings). We will test this using six commonly recognized and frequently monitored therapy-effect pairs. (H2) Knowledge of the precise timing and determinants of therapy-related adverse events from H1 will enable us to alter end-user ordering behaviors in a manner to (a) decrease unnecessary testing and (b) prevent testing too infrequently to detect adverse events. We will specifically study the six therapy-lab pairs of the first hypothesis using CPOE-based interventions. The "TIME for safe and appropriate testing" project will accomplish the following objectives: Aim 1: Construct a secure and confidential reference database, covering August 1999- July 2003 inclusive, of orders for common therapeutic interventions and reports of common laboratory results for inpatients. Aim 2: Validate the representativeness of the database using sample patient chart audits (1) to ensure that the therapy orders stored in the database are a reasonable proxy for therapies administered, and (2) to verify that the targeted lab abnormalities reliably reflect adverse drug effects rather than other abnormalities in patients given the drugs of interest. Aim 3: Demonstrate database utility to define incidence, severity, time course, and covariate determinants of adverse effects for at least 6 out of 11 specified well known, commonly monitored therapy-adverse laboratory result pairs. Aim 4: Perform data mining to discover novel therapeutic toxicities and co-determinants of Aim 3 toxicities. Aim 5: Demonstrate the generic utility of the database for investigating therapy-lab effect pairings, for six of the therapy lab pairs of Aim 3, implement decision support tools via the local CPOE system to (a) decrease unnecessary adverse effect monitoring tests and (b) prevent testing too infrequently to detect adverse events. This will include detailed analysis of the economic impact of at least one such intervention, and construction of a "generic model" for analysis of potential future interventions. Aim 6: Publish/provide access for qualified researchers to an anonymous version of the therapy-lab database as a general research tool to further study adverse effect monitoring strategies.